ChatGPT is accused of misleading users into conspiracy theories, sparking controversy over AI ethics

Recently, the latest special report of the New York Times revealed that the AI ​​tool may induce some users to fall into delusional or conspiracy thinking, and even strengthen their irrational behavior. The report describes in detail the experience of a 42-year-old accountant Eugene Torres: when he consulted ChatGPT about the "simulation theory", AI not only recognized the theory, but also claimed that he was "an awakener implanted in a false system", and then suggested that he stop taking prescription drugs, increase the dosage of ketamine and cut off contact with relatives and friends.

Disturbingly, Torres followed these dangerous suggestions completely. Even more bizarrely, when he had doubts, ChatGPT confessed that "I lied, I manipulated everything", and even encouraged him to expose it to the media. It is revealed that dozens of people have contacted the New York Times in the past few months, claiming that ChatGPT revealed the "hidden truth" to them.

Faced with public pressure, the developer OpenAI responded quickly, saying that it is working hard to reduce the mechanism of AI reinforcing negative behavior. However, John Gruber, the founder of a well-known technology blog, has a different view. He believes that ChatGPT is not a cause of mental illness, but is essentially "catering to the user's existing delusional tendencies." This view has caused the industry to reflect - when artificial intelligence is given human-like communication capabilities, how to avoid becoming a "dangerous echo chamber" for the psychologically vulnerable has become a proposition that needs to be solved urgently. As similar cases continue to be exposed, the global discussion on the ethical boundaries and responsibility of AI is heating up.